Project Summary Studying protein interactions in living cells is crucial for our fundamental understanding of cell function and alterations in disease states. Super-resolution microscopy such as photoactivated localization microscopy (PALM) evolved into a prominent technique to study the co-localization, interaction and function of proteins in cells with unprecedented specificity and resolution. However, a major limitation of PALM is the inability to detect interacting proteins, if one is present at much higher concentration and obscures the signal of the sparsely interacting fraction. This limitation prevents to study the multitude of signaling pathways across all cell lines and disease models whose interacting proteins have concentration differences up to 1,000 fold. This proposal will explore and characterize a novel diffusion-contrast technique for PALM (dcPALM) circumventing this limitation. By combining the high resolution of PALM with a spatially confined photoactivation protocol via a digital mirror array, the different mobilities of interacting and non-interacting proteins will be exploited to yield unprecedented contrast for detecting sparsely interacting proteins. First, the photoactivatable proteins (PAFPs) such as mEos2 fused to a protein of interest will be activated in a small volume of the cell. After a variable lag time allowing non-interacting proteins to diffuse away, only the sparsely interacting proteins that remained at the same location will be excited and imaged. Aim 1 will explore and characterize this technique with a specifically designed human HeLa and yeast cell line expressing membrane- interacting mEos2 proteins at low levels and non-interacting mEos2 proteins expressed up to 3,500-fold higher. By systematically varying the experimental parameters such as lag time, the power and area of photoactivation, and the expression levels of the proteins, a proof of concept will be obtained and the applicable parameter space with the associated increase in contrast of dcPALM will be quantified. To broaden the applicability of dcPALM to the multitude of mobile intracellular structures such as vesicles or signaling clusters, aim 2 will develop a feedback mode, which adjusts the photoactivation area in real time. First, a snapshot of the structures will be recorded in a separate fluorescence channel and converted to a binary mask. This mask will be loaded on the digital mirror array to only photoactivate the structures at their current location. After a variable lag time only proteins that interact with the structures will be detected. Repeating this cycle numerous times will ensure that the photoactivation region is automatically adjusted in real time. A proof of concept will be obtained by applying this technique to detect endogenously tagged mTOR on lysosomes in HeLa cells. The successful development of this technique would pave the way for discovering new protein interactions at previously inaccessible concentration ranges and may lead to the detection of alteration in disease states.